delivered the opinion of the Court.
In this medical negligence case, we consider whether the trial court erred in admitting “habit” evidence from medical personnel ,to prove that a patient did not complain of pain on a specific occasion, and that the defendant’s treatment of the patient conformed to his routine practice.
Jane V. Ligón, administrator of the estate of Pearl V. Vaughan, filed a wrongful death action in the trial court against Dr. Girish Purohit and his medical practice, Southside Cardiology Associates, P.C. (collectively, the defendant). Ligón alleged, among other things, that the defendant was negligent in failing to provide a proper diagnosis and treatment of Vaughan’s heart disease.
The following evidence was presented in a jury trial. In May 1995, Vaughan experienced chest pains and received three days of treatment at the Southside Community Hospital (Southside) in Farmville. Five days after her release, she returned to Southside’s Cardiac Diagnostic Unit (CDU) as an outpatient to take a Persantine stress test. In this test, the drug Persantine is administered to place additional stress on the patient’s heart so that abnormalities can be detected and evaluated.
Vaughan’s daughter, Audrey Johnson, took Vaughan to the hospital for the Persantine stress test and remained there during the course of the procedure. Under the standard protocol for this test, Vaughan completed a medical history form in the CDU and an intravenous “saline lock” was placed in her arm. Vaughan then went to the hos*309pital’s Nuclear Medicine Department where she received an injection of a radioactive medicine. A medical technician took photographic images, commonly referred to as a “nuclear scan,” of Vaughan’s resting heart.
After three hours, Vaughan returned to the CDU where another medical technician connected her to vital sign monitors and obtained various electrocardiograms (EKGs). Dr. Purohit supervised the injection of the Persantine and the administration of the stress test. The stress test took 14 minutes to complete. During the stress test, Vaughan experienced tightness and pain in her chest. At Dr. Purohit’s direction, Vaughan was given nitroglycerin, and her chest pain stopped. After Vaughan completed the stress test, the heart monitor and EKG connections were removed and she returned to the Nuclear Medicine Department where a technician took a second nuclear scan of her heart.
Vaughan left the hospital with Johnson and went to Johnson’s home. That night, Vaughan died in her sleep. All three medical experts who testified at trial agreed that Vaughan probably died from an arrhythmia that resulted in cardiac arrest. One of these three witnesses, Dr. James T. Rittelmeyer, a cardiologist, stated that Vaughan also had experienced a “heart attack” during the stress test administered by Dr. Purohit. The other two medical experts disagreed with that conclusion.
Johnson testified that as she waited in the CDU reception area while her mother was undergoing the stress test, she heard Vaughan call her name. Johnson stated that she went to the area where the test was being administered and found Vaughan lying on a gurney, dressed in her own clothes and not connected to any monitors. Johnson said that Vaughan was crying, trembling, and complaining that she could not breathe and that she had pain in her chest and arm.
Johnson testified that Dr. Purohit was standing nearby, along with two female technicians or nurses. Johnson stated that when she asked Dr. Purohit whether Vaughan’s condition was normal, he assured her that it was and said that her mother would be fine once she went home and rested.
Dr. Purohit testified that he had no independent recollection of Vaughan’s condition in the CDU. Debora S. Hurt, the CDU technician who cared for Vaughan, also had no independent recollection of Vaughan. However, Courtney Gates, the nuclear technologist who obtained the final nuclear scan of Vaughan’s heart after the stress test, testified that she remembered Vaughan. Gates stated that Vaughan complained of indigestion or “stomach upset” at that time, *310but that she never complained of chest pain. At trial, all three medical experts testified that a violation of the standard of care would occur if a patient, complaining of chest and arm pain under the circumstances described by Johnson, were released from the hospital without further evaluation. Thus, a critical factual issue in the trial was whether Vaughan complained of chest and arm pain after completing the stress test.
Over Ligon’s objection, the defendant was permitted to present evidence of the routine or “habit” of Dr. Purohit, Hurt, and Gates in responding to other patients who complained of chest pain after completing stress tests. Dr. Purohit testified that he had administered one or two stress tests per day over the last ten years, and that at least a dozen of those patients had complained of chest pain after completing the test and changing into their own clothes. He stated that whenever this occurred, he re-evaluated the patient by obtaining another EKG and performing a physical examination. Dr. Purohit testified that he had never failed to re-evaluate a patient who complained of chest pain on completion of a stress test.
Hurt testified that she had worked as a cardiac diagnostic technician for ten years. When asked how many times she had observed patients develop complaints similar to those described by Johnson, Hurt responded that such complaints had occurred more than ten times. She testified that when these complaints were brought to her attention, she reconnected the patients to an EKG monitor and had them re-evaluated by a physician.
Gates testified that during the 30 years she had worked as a nuclear technologist, patients had complained of chest pain “more than ten” times. Gates stated that she immediately responded to those complaints by requesting assistance from the cardiac unit or the emergency room.
At the conclusion of the evidence, the jury returned a verdict in favor of the defendant. The trial court entered judgment in accordance with the jury’s verdict.
On appeal, Ligon argues that the trial court erred in admitting the defendant’s “habit” evidence. Ligon asserts that the challenged testimony permitted the jury to speculate that because Dr. Purohit, Hurt, and Gates provided proper medical care to other patients, they provided the same care to Vaughan. Ligon argues that under our holding in Jackson v. Chesapeake & Ohio Ry. Co., 179 Va. 642, 20 S.E.2d 489 (1942), evidence of a defendant’s habitual conduct is inadmissible to prove that the defendant acted in conformance with such conduct on a particular occasion.
*311In response, the defendant argues that the witnesses’ testimony concerning their responses to other patients’ complaints of chest pain was not evidence of general habit such as that addressed in Jackson, but was evidence of “specific responses to a specific situation.” The defendant asserts that in a medical negligence action, when a defendant physician has no memory of a patient, evidence of the physician’s routine or habit is relevant to establish his conduct with regard to that particular patient. The defendant further contends that the challenged testimony, was not offered to prove that the defendant was not negligent, but merely was offered to show that a particular event, Vaughan’s complaint of chest pain, did not occur. We disagree with the defendant’s arguments.
Our decisions do not draw a distinction between “general” and “specific” habit evidence. Instead, the focus of our analysis has been whether the proffered evidence is relevant to the issues at trial. See Cherry v. D.S. Nash Constr. Co., 252 Va. 241, 244-45, 475 S.E.2d 794, 796-97 (1996); Spurlin v. Richardson, 203 Va. 984, 989-90, 128 S.E.2d 273, 277-78 (1962); Jackson, 179 Va. at 650, 20 S.E.2d at 492.
The reasoning we articulated in Jackson is persuasive in resolving the issue before us. There, a plaintiff brought a negligence action for personal injuries he sustained when the truck in which he was a passenger collided with a train. The engineer in charge of the train’s engine testified that on the day of the accident, as well as on the 6th, 7th, 8th, 9th, 22nd, 23rd, and 24th day of every month, he rang the crossing bell and gave other crossing signals prior to the train’s traversing the crossing. The plaintiff attempted to impeach this testimony with proffered testimony from a witness who would have testified that on the same days in a month other than that in which the accident occurred, the crossing bell was not rung before the train crossed the tracks. 179 Va. at 645-46, 20 S.E.2d at 490.
We held that the trial court did not err in refusing to allow the proffered testimony. We stated that evidence of a person’s general habits is not admissible for the purpose of showing the nature of his conduct on a specific occasion. Id. at 649, 20 S.E.2d at 492. Such evidence of habitual conduct is inadmissible because it consists only of collateral facts, from which no fair inferences can be drawn, and tends to mislead the jury and to divert its attention from the issues before the court. See id. at 648, 20 S.E.2d at 491; Cherry, 252 Va. at 244-45, 475 S.E.2d at 796; Spurlin, 203 Va. at 990, 128 S.E.2d at 278.
*312The reasoning we employed in Jackson was a departure from our prior decisions in Alexandria & F.R.R. Co. v. Herndon, 87 Va. 193, 12 S.E. 289 (1890) and Washington, A. and Mt. V. Ry. Co v. Trimyer, 110 Va. 856, 67 S.E. 531 (1910), in which we approved the admission of evidence that a defendant had acted in an habitually negligent manner prior to the accident at issue. In Herndon, we held that evidence of the habitual stopping place of a train at a location where its rear car had no landing place for exiting passengers was admissible to prove that the train was stopped, rather than in motion, at this location when the plaintiff left the rear car and was injured. 87 Va. at 199, 12 S.E. at 291. In Trimyer, we approved the trial court’s admission of evidence that the defendant railroad company, in violation of its alleged duty, previously had failed to stop its train at the same intersection where the plaintiff allegedly was injured by the defendant’s moving train. 110 Va. at 858-59, 67 S.E. at 532.
After Trimyer, however, we repeatedly have stated that evidence of prior negligent habit is inadmissible to prove the acts of negligence alleged at trial. See Cherry, 252 Va. at 244-45, 475 S.E.2d at 796-97; Jackson, 179 Va. at 649, 20 S.E.2d at 492; Southern Ry. Co. v. Rice’s Adm’x, 115 Va. 235, 248-49, 78 S.E. 592, 595 (1913). Moreover, in these negligence cases, we have rejected the admission of habit evidence offered to prove the issues at trial for the primary reason that such evidence is collateral to the proof of those issues.* See id.
In a negligence action, evidence of habitual conduct is inadmissible to prove conduct at the time of the incident complained of because such evidence is collateral to the issues at trial. Thus, the evidence in question before us was inadmissible because it was collateral to the issues whether this decedent complained of chest pains after her stress test, whether the defendant was negligent in treating this patient at the time of the incident complained of, and whether the alleged acts of negligence were a proximate cause of the dece*313dent’s death. See Cherry, 252 Va. at 244, 475 S.E.2d at 796; Jackson, 179 Va. at 648, 20 S.E.2d at 492.
Acceptance of the contrary position urged by the defendant would result in the admission of irrelevant evidence in a variety of actions. For example, a defendant in an automobile negligence action would be permitted to prove that he obeyed a certain traffic signal at an accident scene by testifying that he complies with that signal on a daily basis when driving at that location. We decline to adopt such a rule because the relevant inquiry in a negligence action is not whether a defendant has a habit of compliance with the type of duty at issue, but whether the defendant breached a specific duty owed to the plaintiff at a particular time.
By our holding in this case, we also decline the defendant’s request that we follow the decisions of other jurisdictions that permit evidence of the habitual conduct of medical personnel for the purpose of proving that the defendant’s conduct on a specific occasion conformed to their routine practice. See, e.g., Bloskas v. Murray, 646 P.2d 907, 911 (Colo. 1982); Crawford v. Fayez, 435 S.E.2d 545, 549-50 (N.C.App. 1993). Those decisions represent a departure from our jurisprudence, and we perceive no benefit from the admission of such evidence to warrant a reversal or curtailment of the basic principles articulated in Jackson.
We also disagree with the defendant’s contention that admission of this type of evidence is necessary to counter a plaintiff’s expert testimony on the applicable standard of care, which is based partly on actions taken by other health care providers under the same circumstances. Both factual and expert testimony in a medical negligence action must be relevant to the incident at issue. The testimony of fact witnesses is relevant to show what actually happened on a particular occasion. The testimony of expert witnesses relates to the same specific incident by establishing a standard of care applicable to the defendant’s actions on that particular occasion and by assessing whether those actions conformed to the established standard of care. In contrast, the evidence improperly admitted by the trial court was relevant only to prove events that occurred on other occasions.
For these reasons, we will reverse the trial court’s judgment and remand the case for a new trial in accordance with the principles expressed in this opinion.
Reversed and remanded.
We also note that Graham v. Commonwealth, 127 Va. 808, 103 S.E. 565 (1920), cited by the defendant, is inapposite to the present case. There, we held that since the defendant on trial for murder had asserted a self-defense claim, alleging that the deceased had used violent, profane language and advanced toward him with a gun, the Commonwealth was entitled to introduce rebuttal evidence that the deceased did not have a habit of swearing. 127 Va. at 824, 103 S.E. at 570. We stated that this evidence was admissible under the same principle that allows the admission of character evidence. Id. Thus, our holding in Graham was limited to the use of a narrow category of rebuttal testimony to a claim of self-defense in a criminal prosecution, and is unrelated to the present issue of the admissibility of habit evidence in a negligence action.